Text

theses

My academic work focuses on banking and financial fragility. A common theme of my research agenda is to study the interaction between the financial crises and the government's policy responses. In particular, I explore how the anticipation of government policy reactions affects the incentives and behavior of bank depositors and other investors. I study this same underlying issue in three distinct settings.
In chapter 1, I study financial fragility in a setting where banking contracts are fixed in nominal terms and all transactions take place using money supplied by the central bank. The existing literature has emphasized that, in such settings, a lender of last resort can effectively prevent self-fulfilling bank runs. I show that, in the event of a crisis, the government will be tempted to intervene ex post to limit the effects of inflation. When depositors anticipate such a reaction, those who have an opportunity to withdraw before the intervention occurs may choose to run on their banks. I show that if the government can commit not to intervene, the efficient allocation will be the unique equilibrium outcome. If the government lacks commitment, however, a self- fulfilling bank run can arise even with nominal banking contracts and a lender of last resort.
In chapter 2, which is joint work with Todd Keister, we ask whether policy makers should be transparent about their plans for dealing with a future financial crisis or there is a role for ambiguity in an optimal policy. We study a modern version of the Diamond and Dybvig (1983) model in which the regulator is able to bail out banks experiencing a loss on their assets. We show that when the regulator has a perfect regulatory power, the optimal policy is fully transparent in the sense that the regulator specifies the full bailout policy in advance. When the financial institutions experience a real loss on their assets, the regulator wants to provide insurance by using some tax revenue to bail out those financial institutions when the regulation is imperfect. However, the anticipation of such a bailout distorts ex ante incentives and leads financial institutions to become more illiquid than is socially desirable. In this environment, the regulator can sometimes improve welfare by introducing ambiguity about its bailout policy. We consider two distinctive forms of ambiguity: one in which the regulator either bails out all banks or none and the other in which announces what fraction of banks will be bailed out in advance but does not say which specific banks will be included. In both cases, we show that the optimal policy involves providing bailouts with some positive probability. In addition, we show that the policy maker should aim to minimize the amount of aggregate uncertainty created by its ambiguous policy.
In chapter 3, I study how the resolution policy for failed institutions affects welfare and the fragility of the banking system. I again study a modern version of the classic model of Diamond and Dybvig (1983), this time adding partial deposit insurance, fiscal policies and bank-specific fundamental uncertainty about the investment return. I ask how the remaining assets of a failed financial institution should be distributed among different creditors in the process of resolution when the government's deposit insurance fund covers some deposits and, therefore, has a claim on the institution's remaining assets. I compare then two cases: one where claims of the uninsured depositors are subordinated to those of the deposit insurance fund; and the other in which claims on the failed institution are paid so that the deposit insurance funds and the uninsured depositors share losses. The latter allows risk sharing by shifting resources from public consumption to provide consumption for the uninsured depositors, which in turn can raise welfare and lower some depositors' incentive to run on their banks. However, this approach can distort ex ante incentives for the banks to increase their short-term liabilities, and therefore, can increase their depositors' incentives to run. Numerical exercises show that sharing losses between the deposit insurance fund and the uninsured depositors often improves welfare and promotes financial stability. I show that an increase in the deposit insurance coverage can sometimes lower welfare and make the banking system more susceptible to a run by uninsured depositors because any losses will be concentrated among smaller group of uninsured depositors.

ETD_10089

Ph.D.

Includes bibliographical references

doi:10.7282/t3-9pre-dh07

ETD doctoral

The author owns the copyright to this work.